The Thermodynamics of Covid
The plot above is the historical record of Covid infections and deaths in the United States provided by the New York Times (through their API). What is most noticeable from the data in the top plot above, is that the reported infections “throb”. This behavior is not specific to the US—many cities and countries have the same dissipative behavior. From the top plot above, we also see that the US is not yet in recovery: the green curve in the last few weeks persistently jumps above the red curve.
The Covid coverage on the New York Times front page has changed from the manner in which they report measurements in the last Covid wave (Jan 2021). While the NYT’s API still supports daily tallies, the front page now only presents a 14-day average. The data reported as a 14-day average effectively removes the “throbbing” observed in the previous wave, the characteristic of any far from equilibrium (non-equilibrium) system. Not only does the averaging veil the actual propagation of the virus in our communities, it also produces a repeating and misleading picture when observed average infections are seen to be going down. For example, the media in New York City today (Jan 18, 2022) is holding out hope that New York City Covid has reached its peak because the average infections are falling. In a dissipative structure, the average must go down in preparation for a potential surge forward.
Let’s look at the state of affairs on March 2nd 2022 in New York City (see below) using the daily death toll data provided by the NYTs. We observe that the characteristic pulses of a Covid wave have attenuated and has entered recovery (below, top plot, bottom right hand: green curve is below the red curve). In addition, the middle plot for infection ratio confirms recovery for NYC, since it is consistently negative for the last two weeks. Many counties in the US are far from recovery.
Covid-19 Geiger Counter
Because of issues with Covid-19 testing, confirmed cases data can be misleading. The percentage of positive results as a percentage of testing is a good metric, but it too has been the subject of “truth decay” in the US. Deaths, however, can be counted with high confidence. Our objective is to use machine learning and deaths data to identify the strength of the infection source in the population, and the infection rate per day given the observed death toll and 3% mortality. The infection energy and the infection ratio metrics are independent of the mortality value. Note that it is not critical that the reported deaths are exactly correct. If our errors in counting COVID deaths is systematic, then the conclusions are the same.
Our results can be used to evaluate the effectiveness of public health measures, and help inform when and how they can be relaxed so that we can return to work, and the company of our extended family and friends.
The science and mathematics used here is nearly identical to the design and use of a Geiger Counter (that indirectly measures the strength of a radioactive source through ionization) and non-equilibrium chemical processes.
New Whitepaper Article: Science Strikes Back at Covid-19
Community Interest
All US Counties - 12/14/2020
New York City Community - 2/21/2021
Los Angeles Community - 2/21/2021
San Francisco Bay Area - 2/7/2021
Detroit Community - 2/7/2021
Global Interest
Additional Technical Information
Key reported metrics are designed to be independent of mortality.
The instrument uses a 3% mortality. Note that if the mortality were reduced by half (to 1.5%), then in order to achieve the death toll we observe, the source strength and infection rate would necessarily double. Therefore reducing the mortality for a given death toll implies more suffering and potential to contract the disease. Because there are variants with different infection rates, we have chosen to use a blended rate of 3% mortality.